summaryrefslogtreecommitdiff
path: root/isort/isort.py
blob: 13dc6d4482b341bb3c6508df10c49653fba01efa (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
"""isort.py.

Exposes a simple library to sort through imports within Python code

usage:
    SortImports(file_name)
or:
    sorted = SortImports(file_contents=file_contents).output

Copyright (C) 2013  Timothy Edmund Crosley

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated
documentation files (the "Software"), to deal in the Software without restriction, including without limitation
the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and
to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or
substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF
CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
OTHER DEALINGS IN THE SOFTWARE.

"""
from __future__ import absolute_import, division, print_function, unicode_literals

import codecs
import copy
import itertools
import os
from collections import namedtuple
from datetime import datetime
from difflib import unified_diff
from sys import path as PYTHONPATH
from sys import stderr, stdout

from natsort import natsorted
from pies.overrides import *

from . import settings

SECTION_NAMES = ("FUTURE", "STDLIB", "THIRDPARTY", "FIRSTPARTY", "LOCALFOLDER")
SECTIONS = namedtuple('Sections', SECTION_NAMES)(*range(len(SECTION_NAMES)))


class SortImports(object):
    incorrectly_sorted = False

    def __init__(self, file_path=None, file_contents=None, write_to_stdout=False, check=False,
                 show_diff=False, settings_path=None, **setting_overrides):

        if not settings_path and file_path:
            settings_path = os.path.dirname(os.path.abspath(file_path))
        settings_path = settings_path or os.getcwd()

        self.config = settings.from_path(settings_path).copy()
        for key, value in itemsview(setting_overrides):
            access_key = key.replace('not_', '').lower()
            if type(self.config.get(access_key)) in (list, tuple):
                if key.startswith('not_'):
                    self.config[access_key] = list(set(self.config[access_key]).difference(value))
                else:
                    self.config[access_key] = list(set(self.config[access_key]).union(value))
            else:
                self.config[key] = value

        indent = str(self.config['indent'])
        if indent.isdigit():
            indent = " " * int(indent)
        else:
            indent = indent.strip("'").strip('"')
            if indent.lower() == "tab":
                indent = "\t"
        self.config['indent'] = indent

        self.place_imports = {}
        self.import_placements = {}
        self.remove_imports = [self._format_simplified(removal) for removal in self.config.get('remove_imports', [])]
        self.add_imports = [self._format_natural(addition) for addition in self.config.get('add_imports', [])]
        self._section_comments = ["# " + value for key, value in itemsview(self.config) if
                                  key.startswith('import_heading') and value]

        file_name = file_path
        self.file_path = file_path or ""
        if file_path and not file_contents:
            file_path = os.path.abspath(file_path)
            if self._should_skip(file_path):
                if self.config['verbose']:
                    print("WARNING: {0} was skipped as it's listed in 'skip' setting".format(file_path))
                file_contents = None
            else:
                self.file_path = file_path
                with open(file_path) as file_to_import_sort:
                    file_contents = file_to_import_sort.read()
                    file_contents = PY2 and file_contents.decode('utf8') or file_contents

        if file_contents is None or ("isort:" + "skip_file") in file_contents:
            return

        self.in_lines = file_contents.split("\n")
        self.original_length = len(self.in_lines)
        if (self.original_length > 1 or self.in_lines[:1] not in ([], [""])) or self.config.get('force_adds', False):
            for add_import in self.add_imports:
                self.in_lines.append(add_import)
        self.number_of_lines = len(self.in_lines)

        self.out_lines = []
        self.comments = {'from': {}, 'straight': {}, 'nested': {}, 'above': {'straight': {}, 'from': {}}}
        self.imports = {}
        self.as_map = {}
        for section in itertools.chain(SECTIONS, self.config['forced_separate']):
            self.imports[section] = {'straight': set(), 'from': {}}

        self.index = 0
        self.import_index = -1
        self._first_comment_index_start = -1
        self._first_comment_index_end = -1
        self._parse()
        if self.import_index != -1:
            self._add_formatted_imports()

        self.length_change = len(self.out_lines) - self.original_length
        while self.out_lines and self.out_lines[-1].strip() == "":
            self.out_lines.pop(-1)
        self.out_lines.append("")

        self.output = "\n".join(self.out_lines)
        if self.config.get('atomic', False):
            try:
                compile(self._strip_top_comments(self.out_lines), self.file_path, 'exec', 0, 1)
            except SyntaxError:
                self.output = file_contents
                self.incorrectly_sorted = True
                try:
                    compile(self._strip_top_comments(self.in_lines), self.file_path, 'exec', 0, 1)
                    print("ERROR: {0} isort would have introduced syntax errors, please report to the project!". \
                          format(self.file_path))
                except SyntaxError:
                    print("ERROR: {0} File contains syntax errors.".format(self.file_path))

                return
        if check:
            if self.output == file_contents:
                if self.config['verbose']:
                    print("SUCCESS: {0} Everything Looks Good!".format(self.file_path))
            else:
                print("ERROR: {0} Imports are incorrectly sorted.".format(self.file_path))
                self.incorrectly_sorted = True
                if show_diff or self.config.get('show_diff', False) is True:
                    self._show_diff(file_contents)
            return

        if show_diff or self.config.get('show_diff', False) is True:
            self._show_diff(file_contents)
        elif write_to_stdout:
            stdout.write(self.output)
        elif file_name:
            with codecs.open(self.file_path, encoding='utf-8', mode='w') as output_file:
                output_file.write(self.output)

    def _show_diff(self, file_contents):
        for line in unified_diff(
            file_contents.splitlines(1),
            self.output.splitlines(1),
            fromfile=self.file_path + ':before',
            tofile=self.file_path + ':after',
            fromfiledate=datetime.fromtimestamp(os.path.getmtime(self.file_path)),
            tofiledate=datetime.now()
        ):
            stdout.write(line)

    @staticmethod
    def _strip_top_comments(lines):
        """Strips # comments that exist at the top of the given lines"""
        lines = copy.copy(lines)
        while lines and lines[0].startswith("#"):
            lines = lines[1:]
        return "\n".join(lines)

    def _should_skip(self, filename):
        """Returns True if the file should be skipped based on the loaded settings."""
        if filename in self.config['skip']:
            return True

        position = os.path.split(filename)
        while position[1]:
            if position[1] in self.config['skip']:
                return True
            position = os.path.split(position[0])

    def place_module(self, moduleName):
        """Tries to determine if a module is a python std import, third party import, or project code:

        if it can't determine - it assumes it is project code

        """
        for forced_separate in self.config['forced_separate']:
            if moduleName.startswith(forced_separate) or moduleName.startswith("." + forced_separate):
                return forced_separate

        if moduleName.startswith("."):
            return SECTIONS.LOCALFOLDER

        # Try to find most specific placement instruction match (if any)
        parts = moduleName.split('.')
        module_names_to_check = ['.'.join(parts[:first_k]) for first_k in range(len(parts), 0, -1)]
        for module_name_to_check in module_names_to_check:
            for placement, config_key in (
                    (SECTIONS.FUTURE, 'known_future_library'),
                    (SECTIONS.STDLIB, 'known_standard_library'),
                    (SECTIONS.THIRDPARTY, 'known_third_party'),
                    (SECTIONS.FIRSTPARTY, 'known_first_party'),
                    ):
                if module_name_to_check in self.config[config_key]:
                    return placement

        paths = PYTHONPATH
        virtual_env = os.environ.get('VIRTUAL_ENV', None)
        if virtual_env:
            paths = list(paths)
            for version in ((2, 6), (2, 7), (3, 0), (3, 1), (3, 2), (3, 3), (3, 4)):
                paths.append("{0}/lib/python{1}.{2}/site-packages".format(virtual_env, version[0], version[1]))

        for prefix in paths:
            module_path = "/".join((prefix, moduleName.replace(".", "/")))
            package_path = "/".join((prefix, moduleName.split(".")[0]))
            if (os.path.exists(module_path + ".py") or os.path.exists(module_path + ".so") or
               (os.path.exists(package_path) and os.path.isdir(package_path))):
                if "site-packages" in prefix or "dist-packages" in prefix:
                    return SECTIONS.THIRDPARTY
                elif "python2" in prefix.lower() or "python3" in prefix.lower():
                    return SECTIONS.STDLIB
                else:
                    return SECTIONS.FIRSTPARTY

        return SECTION_NAMES.index(self.config['default_section'])

    def _get_line(self):
        """Returns the current line from the file while incrementing the index."""
        line = self.in_lines[self.index]
        self.index += 1
        return line

    @staticmethod
    def _import_type(line):
        """If the current line is an import line it will return its type (from or straight)"""
        if "isort:skip" in line:
            return
        elif line.startswith('import '):
            return "straight"
        elif line.startswith('from '):
            return "from"

    def _at_end(self):
        """returns True if we are at the end of the file."""
        return self.index == self.number_of_lines

    @staticmethod
    def _module_key(module_name, config, sub_imports=False):
        prefix = ""
        module_name = str(module_name)
        if sub_imports and config['order_by_type']:
            if module_name.isupper():
                prefix = "A"
            elif module_name[0:1].isupper():
                prefix = "B"
            else:
                prefix = "C"
        module_name = module_name.lower()
        return "{0}{1}{2}".format(module_name in config['force_to_top'] and "A" or "B", prefix,
                                  config['length_sort'] and (str(len(module_name)) + ":" + module_name) or module_name)

    def _add_comments(self, comments, original_string=""):
        """
            Returns a string with comments added
        """
        return comments and "{0}  # {1}".format(self._strip_comments(original_string)[0],
                                                "; ".join(comments)) or original_string

    def _wrap(self, line):
        """
            Returns an import wrapped to the specified line-length, if possible.
        """
        if len(line) > self.config['line_length']:
            for splitter in ("import", "."):
                if splitter in line and not line.strip().startswith(splitter):
                    line_parts = line.split(splitter)
                    next_line = []
                    while (len(line) + 2) > (self.config['wrap_length'] or self.config['line_length']) and line_parts:
                        next_line.append(line_parts.pop())
                        line = splitter.join(line_parts)
                    if not line:
                        line = next_line.pop()
                    return "{0}{1} \\\n{2}".format(line, splitter,
                                                  self._wrap(self.config['indent'] + splitter.join(next_line).lstrip()))

        return line

    def _add_straight_imports(self, straight_modules, section, section_output):
        for module in straight_modules:
            if module in self.remove_imports:
                continue

            if module in self.as_map:
                import_definition = "import {0} as {1}".format(module, self.as_map[module])
            else:
                import_definition = "import {0}".format(module)

            comments_above = self.comments['above']['straight'].get(module, None)
            if comments_above:
                section_output.append(comments_above)
            section_output.append(self._add_comments(self.comments['straight'].get(module), import_definition))

    def _add_from_imports(self, from_modules, section, section_output):
        for module in from_modules:
            if module in self.remove_imports:
                continue

            import_start = "from {0} import ".format(module)
            from_imports = list(self.imports[section]['from'][module])
            from_imports = natsorted(from_imports, key=lambda key: self._module_key(key, self.config, True))
            if self.remove_imports:
                from_imports = [line for line in from_imports if not "{0}.{1}".format(module, line) in
                                self.remove_imports]

            for from_import in copy.copy(from_imports):
                submodule = module + "." + from_import
                import_as = self.as_map.get(submodule, False)
                if import_as:
                    import_definition = "{0} as {1}".format(from_import, import_as)
                    if self.config['combine_as_imports'] and not ("*" in from_imports and
                                                                    self.config['combine_star']):
                        from_imports[from_imports.index(from_import)] = import_definition
                    else:
                        import_statement = self._wrap(import_start + import_definition)
                        comments = self.comments['straight'].get(submodule)
                        import_statement = self._add_comments(comments, import_statement)
                        section_output.append(import_statement)
                        from_imports.remove(from_import)

            if from_imports:
                comments = self.comments['from'].get(module)
                if "*" in from_imports and self.config['combine_star']:
                    import_statement = self._wrap(self._add_comments(comments, "{0}*".format(import_start)))
                elif self.config['force_single_line']:
                    import_statements = []
                    for from_import in from_imports:
                        single_import_line = self._add_comments(comments, import_start + from_import)
                        comment = self.comments['nested'].get(module, {}).get(from_import, None)
                        if comment:
                            single_import_line += "{0} {1}".format(comments and ";" or "  #", comment)
                        import_statements.append(self._wrap(single_import_line))
                        comments = None
                    import_statement = "\n".join(import_statements)
                else:
                    star_import = False
                    if "*" in from_imports:
                        section_output.append(self._add_comments(comments, "{0}*".format(import_start)))
                        from_imports.remove('*')
                        star_import = True
                        comments = None

                    for from_import in copy.copy(from_imports):
                        comment = self.comments['nested'].get(module, {}).get(from_import, None)
                        if comment:
                            single_import_line = self._add_comments(comments, import_start + from_import)
                            single_import_line += "{0} {1}".format(comments and ";" or "  #", comment)
                            above_comments = self.comments['above']['from'].get(module, None)
                            if above_comments:
                                section_output.extend(above_comments)
                            section_output.append(self._wrap(single_import_line))
                            from_imports.remove(from_import)
                            comments = None

                    if star_import:
                        import_statement = import_start + (", ").join(from_imports)
                    else:
                        import_statement = self._add_comments(comments, import_start + (", ").join(from_imports))
                    if not from_imports:
                        import_statement = ""
                    if len(import_statement) > self.config['line_length']:
                        import_statement = self._wrap(import_statement)
                    if len(from_imports) > 1 and (
                        len(import_statement) > self.config['line_length']
                        or self.config.get('force_from_wrap')
                    ):
                        output_mode = settings.WrapModes._fields[self.config.get('multi_line_output',
                                                                                    0)].lower()
                        formatter = getattr(self, "_output_" + output_mode, self._output_grid)
                        dynamic_indent = " " * (len(import_start) + 1)
                        indent = self.config['indent']
                        line_length = self.config['wrap_length'] or self.config['line_length']
                        import_statement = formatter(import_start, copy.copy(from_imports),
                                                    dynamic_indent, indent, line_length, comments)
                        if self.config['balanced_wrapping']:
                            lines = import_statement.split("\n")
                            line_count = len(lines)
                            if len(lines) > 1:
                                minimum_length = min([len(line) for line in lines[:-1]])
                            else:
                                minimum_length = 0
                            new_import_statement = import_statement
                            while (len(lines[-1]) < minimum_length and
                                    len(lines) == line_count and line_length > 10):
                                import_statement = new_import_statement
                                line_length -= 1
                                new_import_statement = formatter(import_start, copy.copy(from_imports),
                                                                dynamic_indent, indent, line_length, comments)
                                lines = new_import_statement.split("\n")

                if import_statement:
                    above_comments = self.comments['above']['from'].get(module, None)
                    if above_comments:
                        section_output.extend(above_comments)
                    section_output.append(import_statement)

    def _add_formatted_imports(self):
        """Adds the imports back to the file.

        (at the index of the first import) sorted alphabetically and split between groups

        """
        output = []
        for section in itertools.chain(SECTIONS, self.config['forced_separate']):
            straight_modules = list(self.imports[section]['straight'])
            straight_modules = natsorted(straight_modules, key=lambda key: self._module_key(key, self.config))
            from_modules = sorted(list(self.imports[section]['from'].keys()))
            from_modules = natsorted(from_modules, key=lambda key: self._module_key(key, self.config, ))

            section_output = []
            if self.config.get('from_first', False):
                self._add_from_imports(from_modules, section, section_output)
                self._add_straight_imports(straight_modules, section, section_output)
            else:
                self._add_straight_imports(straight_modules, section, section_output)
                self._add_from_imports(from_modules, section, section_output)

            if section_output:
                section_name = section
                if section in SECTIONS:
                    section_name = SECTION_NAMES[section]
                if section_name in self.place_imports:
                    self.place_imports[section_name] = section_output
                    continue

                section_title = self.config.get('import_heading_' + str(section_name).lower(), '')
                if section_title:
                    section_output.insert(0, "# " + section_title)
                output += section_output + ['']

        while [character.strip() for character in output[-1:]] == [""]:
            output.pop()

        output_at = 0
        if self.import_index < self.original_length:
            output_at = self.import_index
        elif self._first_comment_index_end != -1 and self._first_comment_index_start <= 2:
            output_at = self._first_comment_index_end
        self.out_lines[output_at:0] = output

        imports_tail = output_at + len(output)
        while [character.strip() for character in self.out_lines[imports_tail: imports_tail + 1]] == [""]:
            self.out_lines.pop(imports_tail)

        if len(self.out_lines) > imports_tail:
            next_construct = ""
            self._in_quote = False
            for line in self.out_lines[imports_tail:]:
                if not self._skip_line(line) and not line.strip().startswith("#") and line.strip():
                    next_construct = line
                    break

            if self.config['lines_after_imports'] != -1:
                self.out_lines[imports_tail:0] = ["" for line in range(self.config['lines_after_imports'])]
            elif next_construct.startswith("def") or next_construct.startswith("class") or \
               next_construct.startswith("@"):
                self.out_lines[imports_tail:0] = ["", ""]
            else:
                self.out_lines[imports_tail:0] = [""]

        if self.place_imports:
            new_out_lines = []
            for index, line in enumerate(self.out_lines):
                new_out_lines.append(line)
                if line in self.import_placements:
                    new_out_lines.extend(self.place_imports[self.import_placements[line]])
                    if len(self.out_lines) <= index or self.out_lines[index + 1].strip() != "":
                        new_out_lines.append("")
            self.out_lines = new_out_lines


    def _output_grid(self, statement, imports, white_space, indent, line_length, comments):
        statement += "(" + imports.pop(0)
        while imports:
            next_import = imports.pop(0)
            next_statement = self._add_comments(comments, statement + ", " + next_import)
            if len(next_statement.split("\n")[-1]) + 1 > line_length:
                statement = (self._add_comments(comments, "{0},".format(statement)) +
                             "\n{0}{1}".format(white_space, next_import))
                comments = None
            else:
                statement += ", " + next_import
        return statement + ("," if self.config['include_trailing_comma'] else "") + ")"

    def _output_vertical(self, statement, imports, white_space, indent, line_length, comments):
        first_import = self._add_comments(comments, imports.pop(0) + ",") + "\n" + white_space
        return "{0}({1}{2}{3})".format(
            statement,
            first_import,
            (",\n" + white_space).join(imports),
            "," if self.config['include_trailing_comma'] else "",
        )

    def _output_hanging_indent(self, statement, imports, white_space, indent, line_length, comments):
        statement += imports.pop(0)
        while imports:
            next_import = imports.pop(0)
            next_statement = self._add_comments(comments, statement + ", " + next_import)
            if len(next_statement.split("\n")[-1]) + 3 > line_length:
                next_statement = (self._add_comments(comments, "{0}, \\".format(statement)) +
                                  "\n{0}{1}".format(indent, next_import))
                comments = None
            statement = next_statement
        return statement

    def _output_vertical_hanging_indent(self, statement, imports, white_space, indent, line_length, comments):
        return "{0}({1}\n{2}{3}{4}\n)".format(
            statement,
            self._add_comments(comments),
            indent,
            (",\n" + indent).join(imports),
            "," if self.config['include_trailing_comma'] else "",
         )

    def _output_vertical_grid_common(self, statement, imports, white_space, indent, line_length, comments):
        statement += self._add_comments(comments, "(") + "\n" + indent + imports.pop(0)
        while imports:
            next_import = imports.pop(0)
            next_statement = "{0}, {1}".format(statement, next_import)
            if len(next_statement.split("\n")[-1]) + 1 > line_length:
                next_statement = "{0},\n{1}{2}".format(statement, indent, next_import)
            statement = next_statement
        if self.config['include_trailing_comma']:
            statement += ','
        return statement

    def _output_vertical_grid(self, statement, imports, white_space, indent, line_length, comments):
        return self._output_vertical_grid_common(statement, imports, white_space, indent, line_length, comments) + ")"

    def _output_vertical_grid_grouped(self, statement, imports, white_space, indent, line_length, comments):
        return self._output_vertical_grid_common(statement, imports, white_space, indent, line_length, comments) + "\n)"

    @staticmethod
    def _strip_comments(line, comments=None):
        """Removes comments from import line."""
        if comments is None:
            comments = []

        new_comments = False
        comment_start = line.find("#")
        if comment_start != -1:
            comments.append(line[comment_start + 1:].strip())
            new_comments = True
            line = line[:comment_start]

        return line, comments, new_comments

    @staticmethod
    def _format_simplified(import_line):
        import_line = import_line.strip()
        if import_line.startswith("from "):
            import_line = import_line.replace("from ", "")
            import_line = import_line.replace(" import ", ".")
        elif import_line.startswith("import "):
            import_line = import_line.replace("import ", "")

        return import_line

    @staticmethod
    def _format_natural(import_line):
        import_line = import_line.strip()
        if not import_line.startswith("from ") and not import_line.startswith("import "):
            if not "." in import_line:
                return "import {0}".format(import_line)
            parts = import_line.split(".")
            end = parts.pop(-1)
            return "from {0} import {1}".format(".".join(parts), end)

        return import_line

    def _skip_line(self, line):
        skip_line = self._in_quote
        if '"' in line or "'" in line:
            index = 0
            if self._first_comment_index_start == -1:
                self._first_comment_index_start = self.index
            while index < len(line):
                if line[index] == "\\":
                    index += 1
                elif self._in_quote:
                    if line[index:index + len(self._in_quote)] == self._in_quote:
                        self._in_quote = False
                        if self._first_comment_index_end == -1:
                            self._first_comment_index_end = self.index
                elif line[index] in ("'", '"'):
                    long_quote = line[index:index + 3]
                    if long_quote in ('"""', "'''"):
                        self._in_quote = long_quote
                        index += 2
                    else:
                        self._in_quote = line[index]
                elif line[index] == "#":
                    break
                index += 1

        return skip_line or self._in_quote

    def _strip_syntax(self, import_string):
        import_string = import_string.replace("_import", "[[i]]")
        for remove_syntax in ['\\', '(', ')', ',']:
            import_string = import_string.replace(remove_syntax, " ")
        import_list = import_string.split()
        for key in ('from', 'import'):
            if key in import_list:
                import_list.remove(key)
        import_string = ' '.join(import_list)
        import_string = import_string.replace("[[i]]", "_import")
        return import_string

    def _parse(self):
        """Parses a python file taking out and categorizing imports."""
        self._in_quote = False
        while not self._at_end():
            line = self._get_line()
            skip_line = self._skip_line(line)

            if line in self._section_comments and not skip_line:
                if self.import_index == -1:
                    self.import_index = self.index - 1
                continue

            if "isort:" + "imports-" in line and line.startswith("#"):
                section = line.split("isort:" + "imports-")[-1].split()[0]
                self.place_imports[section.upper()] = []
                self.import_placements[line] = section.upper()

            if ";" in line:
                for part in (part.strip() for part in line.split(";")):
                    if part and not part.startswith("from ") and not part.startswith("import "):
                        skip_line = True

            import_type = self._import_type(line)
            if not import_type or skip_line:
                self.out_lines.append(line)
                continue

            for line in (line.strip() for line in line.split(";")):
                import_type = self._import_type(line)
                if not import_type:
                    self.out_lines.append(line)
                    continue

                line = line.replace("\t", " ")
                if self.import_index == -1:
                    self.import_index = self.index - 1

                nested_comments = {}
                import_string, comments, new_comments = self._strip_comments(line)
                stripped_line = [part for part in self._strip_syntax(import_string).strip().split(" ") if part]

                if import_type == "from" and len(stripped_line) == 2 and stripped_line[1] != "*" and new_comments:
                    nested_comments[stripped_line[-1]] = comments[0]

                if "(" in line and not self._at_end():
                    while not line.strip().endswith(")") and not self._at_end():
                        line, comments, new_comments = self._strip_comments(self._get_line(), comments)
                        stripped_line = self._strip_syntax(line).strip()
                        if import_type == "from" and stripped_line and not " " in stripped_line and new_comments:
                            nested_comments[stripped_line] = comments[-1]
                        import_string += "\n" + line
                else:
                    while line.strip().endswith("\\"):
                        line, comments, new_comments = self._strip_comments(self._get_line(), comments)
                        stripped_line = self._strip_syntax(line).strip()
                        if import_type == "from" and stripped_line and not " " in stripped_line and new_comments:
                            nested_comments[stripped_line] = comments[-1]
                        if import_string.strip().endswith(" import") or line.strip().startswith("import "):
                            import_string += "\n" + line
                        else:
                            import_string = import_string.rstrip().rstrip("\\") + line.lstrip()

                if import_type == "from":
                    parts = import_string.split(" import ")
                    from_import = parts[0].split(" ")
                    import_string = " import ".join([from_import[0] + " " + "".join(from_import[1:])] + parts[1:])

                imports = self._strip_syntax(import_string).split()
                if "as" in imports and (imports.index('as') + 1) < len(imports):
                    while "as" in imports:
                        index = imports.index('as')
                        if import_type == "from":
                            module = imports[0] + "." + imports[index - 1]
                            self.as_map[module] = imports[index + 1]
                        else:
                            module = imports[index - 1]
                            self.as_map[module] = imports[index + 1]
                        if not self.config['combine_as_imports']:
                            self.comments['straight'][module] = comments
                            comments = []
                        del imports[index:index + 2]
                if import_type == "from":
                    import_from = imports.pop(0)
                    root = self.imports[self.place_module(import_from)][import_type]
                    for import_name in imports:
                        associated_commment = nested_comments.get(import_name)
                        if associated_commment:
                            self.comments['nested'].setdefault(import_from, {})[import_name] = associated_commment
                            comments.pop(comments.index(associated_commment))
                    if comments:
                        self.comments['from'].setdefault(import_from, []).extend(comments)

                    if len(self.out_lines) > self.import_index:
                        last = self.out_lines and self.out_lines[-1].rstrip() or ""
                        while last.startswith("#") and not last.endswith('"""') and not last.endswith("'''"):
                            self.comments['above']['from'].setdefault(import_from, []).insert(0, self.out_lines.pop(-1))
                            last = self.out_lines and self.out_lines[-1].rstrip() or ""

                    if root.get(import_from, False):
                        root[import_from].update(imports)
                    else:
                        root[import_from] = set(imports)
                else:
                    for module in imports:
                        if comments:
                            self.comments['straight'][module] = comments
                            comments = None

                        if len(self.out_lines) > self.import_index:
                            last = self.out_lines and self.out_lines[-1].rstrip() or ""
                            while last.startswith("#") and not last.endswith('"""') and not last.endswith("'''"):
                                self.comments['above']['from'].setdefault(module, []).insert(0, self.out_lines.pop(-1))
                                last = self.out_lines and self.out_lines[-1].rstrip() or ""
                        self.imports[self.place_module(module)][import_type].add(module)